_ArrayAdd problem

[Melba23]

thebreaker,

Here is a modified _ArrayAdd function where if $vValue is an array and $vDataType is the string "array" then for a 1D array only it is added as a single element:

<snip>

I am not adding the functionality for 2D arrays - too difficult to decide the element in which to add the array.

What do you think? Please test it for me and if it all looks good I will add it to the next Beta.

M23

Edited July 13, 2014 by Melba23
Code removed - see below for newer version

[DerPensionist]

This code seems not to work as designed.

first example returns 0 and adds the new array  entries  1-3

second example returns 3 (old Ubound) and no changes of array.

expandcollapse popup

#include <Array.au3>

Global $aArray_1[3] = [1, 2, 3]
Global $aArray_2[3] = [4, 5, 6]
Global $aContainer[0]

$rc1=_ArrayAdd_Mod($aContainer, $aArray_1, Default, Default, Default, "Array") ; Add as a single element
$er1=@error
$rc2=_ArrayAdd_Mod($aContainer, $aArray_2)                                     ; Add as separate elements
$er2=@error
; See results
_ArrayDisplay($aContainer, "Container="& $rc1 &"["&$er1&"]", Default, 8)
_ArrayDisplay($aContainer[0], "Array= "& $rc2 &"["&$er2&"]", Default, 8)

; If $vValue is an array and $vDataType = "array" - then for a 1D array it is added as a single element
Func _ArrayAdd_Mod(ByRef $avArray, $vValue, $iStart = 0, $sDelim_Item = "|", $sDelim_Row = @CRLF, $vDataType = 0)

    If $iStart = Default Then $iStart = 0
    If $sDelim_Item = Default Then $sDelim_Item = "|"
    If $sDelim_Row = Default Then $sDelim_Row = @CRLF
    If $vDataType = Default Then $vDataType = 0
    If Not IsArray($avArray) Then Return SetError(1, 0, -1)
    Local $iDim_1 = UBound($avArray, $UBOUND_ROWS)
    Switch UBound($avArray, $UBOUND_DIMENSIONS)
        Case 1
            If IsArray($vValue) Then
                If UBound($vValue, $UBOUND_DIMENSIONS) <> 1 Then Return SetError(5, 0, -1)
                If String($vDataType) <> "ARRAY" Then
                    $vDataType = 0
                EndIf
            Else
                Local $aTmp = StringSplit($vValue, $sDelim_Item, $STR_NOCOUNT + $STR_ENTIRESPLIT)
                If UBound($aTmp, $UBOUND_ROWS) = 1 Then
                    $aTmp[0] = $vValue
                    $vDataType = 0
                EndIf
                $vValue = $aTmp
            EndIf
            If String($vDataType) = "ARRAY" Then
                ReDim $avArray[$iDim_1 + 1]
                $avArray[$iDim_1] = $vValue
                Return $iDim_1
            Else
                Local $iAdd = UBound($vValue, $UBOUND_ROWS)
                ReDim $avArray[$iDim_1 + $iAdd]
                For $i = 0 To $iAdd - 1
                    If IsFunc($vDataType) Then
                        $avArray[$iDim_1 + $i] = $vDataType($vValue[$i])
                    Else
                        $avArray[$iDim_1 + $i] = $vValue[$i]
                    EndIf
                Next
                Return $iDim_1 + $iAdd - 1
            EndIf
        Case 2
            Local $iDim_2 = UBound($avArray, $UBOUND_COLUMNS)
            If $iStart < 0 Or $iStart > $iDim_2 - 1 Then Return SetError(4, 0, -1)
            Local $iValDim_1, $iValDim_2
            If IsArray($vValue) Then
                If UBound($vValue, $UBOUND_DIMENSIONS) <> 2 Then Return SetError(5, 0, -1)
                $iValDim_1 = UBound($vValue, $UBOUND_ROWS)
                $iValDim_2 = UBound($vValue, $UBOUND_COLUMNS)
                $vDataType = 0
            Else
                ; Convert string to 2D array
                Local $aSplit_1 = StringSplit($vValue, $sDelim_Row, $STR_NOCOUNT + $STR_ENTIRESPLIT)
                $iValDim_1 = UBound($aSplit_1, $UBOUND_ROWS)
                StringReplace($aSplit_1[0], $sDelim_Item, "")
                $iValDim_2 = @extended + 1
                Local $aTmp[$iValDim_1][$iValDim_2], $aSplit_2
                For $i = 0 To $iValDim_1 - 1
                    $aSplit_2 = StringSplit($aSplit_1[$i], $sDelim_Item, $STR_NOCOUNT + $STR_ENTIRESPLIT)
                    For $j = 0 To $iValDim_2 - 1
                        $aTmp[$i][$j] = $aSplit_2[$j]
                    Next
                Next
                $vValue = $aTmp
            EndIf
            ; Check if too many columns to fit
            If UBound($vValue, $UBOUND_COLUMNS) + $iStart > UBound($avArray, $UBOUND_COLUMNS) Then Return SetError(3, 0, -1)
            ReDim $avArray[$iDim_1 + $iValDim_1][$iDim_2]
            For $iWriteTo_Index = 0 To $iValDim_1 - 1
                For $j = 0 To $iDim_2 - 1
                    If $j < $iStart Then
                        $avArray[$iWriteTo_Index + $iDim_1][$j] = ""
                    ElseIf $j - $iStart > $iValDim_2 - 1 Then
                        $avArray[$iWriteTo_Index + $iDim_1][$j] = ""
                    Else
                        If IsFunc($vDataType) Then
                            $avArray[$iWriteTo_Index + $iDim_1][$j] = $vDataType($vValue[$iWriteTo_Index][$j - $iStart])
                        Else
                            $avArray[$iWriteTo_Index + $iDim_1][$j] = $vValue[$iWriteTo_Index][$j - $iStart]
                        EndIf
                    EndIf
                Next
            Next
        Case Else
            Return SetError(2, 0, -1)
    EndSwitch

    Return UBound($avArray, $UBOUND_ROWS) - 1

EndFunc   ;==> _ArrayAdd_Mod

on the otherhand literal "Array" as keyword to use an array as string ?

I beleave "String" would be more descriptive.

Edited July 13, 2014 by DerPensionist

[Melba23]

DerPensionist,

The code works exactly as designed. As explained in the Help file:

 

Return Value: Success: the index of last added item

So what you are seeing is:

First call: Entire array added to the [0] element of the container - 0 returned as expected

Second call: Array elements added as separate container elements (1 to 3) - 3 returned as expected

Exactly as the Help file describes.

As for using a string as the datatype parameter - I am working on a newer version where that parameter will always be a string defining the required datatype and the function will do the magic internally. I am still testing but the idea would be:

 

If $sDataType = "Array" and $aArray is 1D

- If $vValue is an array it is added as a single element with internal elements retaining existing datatypes

- If $vValue is not an array it is added as separate elements (if in a delimited strings added as strings)

If $sDataType = "Int", "Number", "String", HWnd", "Ptr"

- If $vValue is not an array then all items converted to that datatype

- Ignored if $vValue is an array

If $sDataType is not one of the above then it is ignored

Then the user will not get confused about whether to use a function or a string - I think it is simpler that way.

M23

[czardas]

Regarding the delimiters, why not search the input as default behaviour: As soon as you find a private range unicode character not appearing in the input: wham bam! That's how I do it in my own projects. The user will still need to specify delimiter positions*, or choose a user defined option ==> "|".

* Hmm (i realise it doesn't quite work in this case)! Even so, using a default delimiter that you are actually able to type seems risky.

Edited July 13, 2014 by czardas

[Melba23]

czardas,

AutoIt used predefined delimiters in many other functions (GUICtrlSetData & GUICtrlListViewCreateItem spring immediately to mind). To my mind, forcing the user to define the delimiters each time they use the function seems a much less satisfactory solution that requiring that they only need do so when there is a need to display the default characters.

M23

[czardas]

Maybe! Either way the delimiters have to be positioned in the string by the user. If there was a default keyword Delim or (less good) a Global Const $DELIM variable, which could be identified unambiguously by the interpreter, then there would be no need to know what the delim is. It just would need inserting into the string. That being said, it is more complicated for a beginner although probably more reliable.

I'm sure you've weighed this up already.

Edited July 13, 2014 by czardas

[Melba23]

DerPensionist,

There is no "destroyed element 0" - there is nothing shown in for element [0] in the first _ArrayDisplay dialog because the element contains an array. You can see the content of that array in the second _ArrayDisplay dialog:

_ArrayDisplay($aContainer[0], "Array", Default, 8)

which displays the 3 elements (1, 2, 3) that were added to the container array as an array. Are you telling me that the 2 dialogs show the same thing?

And as some of you seem so determined not to have default delimiters I will look into making them required parameters. But I promise that if I do change anything, I will forward all the subsequent complaints directly to those who asked for the change.

M23

Edit: This is a reply to a now deleted post - so it now makes no sense, or at least less than it did.

Edited July 13, 2014 by Melba23

[DerPensionist]

Sorry i found my misinterpretig.

[Melba23]

DerPensionist,

No problem - glad you got there.

M23

[Melba23]

Hi all,

Here is the latest version of a modified function - using strings to determine any forced datatypes - with some example code to show how it works:

<snip>

The Help file would read as follows:

<snip>

After a lot of thought I have decided to maintain the current default values for the delimiters - most people do not seem to have a problem with it and it would be yet another script-breaking change.

As always, comments on all or any of the above welcome.

M23

Edited July 19, 2014 by Melba23
Code removed

[czardas]

You don't have to change anything for me. I also never use _ArrayAdd(). It is definately better than it was before, as I said previously. I remember having trouble with pipe as a delimiter when I started out with AutoIt. I can't remember exactly why but I found a lot of stuff annoying - much of which wasn't anyone else's fault. My biggest annoyance was how binary is implemented and automatically interpreted to signify something specific - that probably won't change. I know how to fix that now, and it's veering off-topic anyway.

Edited July 13, 2014 by czardas

[guinness]

czardas,

I would really advise that you put your efforts into complaining about those UDF authors who use unnecessary Global variables instead of Global constants. PreExpand has show it's relatively simple to reduce code into an unreadable state, which is why I feel it's time you moved on.

Remember pre-optimisation is the root of all evil.

Edit: I have spent the best part of 30 mins modifying examples in the help file due to adding $STM_GETIMAGE and $STM_SETIMAGE. This is not because I enjoy annoying the community of old AutoIt programmers, but because I want to improve AutoIt's reputation and improve standards.

Edited July 14, 2014 by guinness

[czardas]

guinness,

I try to avoid global variables and prefer to use comments if I pass pre-defined binary variants to a function. I don't quite understand your criticism.

Edited July 15, 2014 by czardas

[guinness]

What you do in your own code is your business, just don't criticise design choices of Devs/MVPs/Mods because it doesn't fit your styling of coding. If there is a valid reason then I am happy to listen as to why we should change this design choice, but so far no one has come up with a valid reason.

This is an open invitation for anyone to PM me as to why constant variables are bad.

Edited July 15, 2014 by guinness

[czardas]

I do criticise some things, but not so much this. The need for you Devs to work as a team is significant factor in how you organize your code, but it doesn't mean everyone has to copy this model exactly. As a learner I feel comfortable having the binary visible. One day I may understand it, but that will never happen if it is concealed behind a substituted name.

[guinness]

OK.

[mLipok]

As a learner I feel comfortable having the binary visible.

 

If I can do this I would like to add my own thoughts.

I am a former student (oh gosh it's almost 10 years ago), I would prefer if it was explained to me by way of a code abstraction and conceptual framework.

If the documentation says:

Can be a combination of the following:
    $FO_READ (0) = Read mode (default)
    $FO_APPEND (1) = Write mode (append to end of file)
    $FO_OVERWRITE (2) = Write mode (erase previous contents)
    $FO_CREATEPATH (8) = Create directory structure if it doesn't exist (See Remarks).
    $FO_BINARY (16) = Force binary mode (See Remarks).
    $FO_UNICODE or $FO_UTF16_LE (32) = Use Unicode UTF16 Little Endian reading and writing mode. Reading does not override existing BOM.
    $FO_UTF16_BE (64) = Use Unicode UTF16 Big Endian reading and writing mode. Reading does not override existing BOM.
    $FO_UTF8 (128) = Use Unicode UTF8 (with BOM) reading and writing mode. Reading does not override existing BOM.
    $FO_UTF8_NOBOM (256) = Use Unicode UTF8 (without BOM) reading and writing mode.
    $FO_UTF8_FULL (16384) = When opening for reading and no BOM is present, use full file UTF8 detection. If this is not used then only the initial part of the file is checked for UTF8.
The folder path must already exist (except using $FO_CREATEPATH mode - See Remarks).
Constants are defined in FileConstants.au3

it is enough to say once at the beginning of the lecture, the importance and value of the constant, and then use it as an abstract notion.

 

And after that it is easier to read and analyze this:

Local $hFileOpen = FileOpen($sFilePath, $FO_READ + $FO_OVERWRITE)

then:

Local $hFileOpen = FileOpen($sFilePath, 0 + 2)

In this way, faster and better develops the mind of the student.

After some time getting used to, reading the code, it reads contextually, and do not need to focus on the meaning of the numbers.

btw.

With one to agree, everyone should at least once to know the value of constants, which is why every time I see the documentation that deficiencies in this area, I try to immediately react by TrackTickets.

ps.

I have a didactical training, but I practice just as much as I lead trainings and presentations with my clients.

Edited July 15, 2014 by mLipok

[czardas]

I like to be frugal with the use of memory and dependancies for several reasons. Comments are non-intrusive and fulfil the same requirements. It's a trade off more than anything else.

[Melba23]

All,

In case anyone still remembers what this thread was about before it was so comprehensively hijacked - this is the last call for comments on the suggested change to the _ArrayAdd function at post #30 above.

M23

[thebreaker]

All,

In case anyone still remembers what this thread was about before it was so comprehensively hijacked - this is the last call for comments on the suggested change to the _ArrayAdd function at post #30 above.

M23

 

I couldn't find the time to test the new function in the setting at work with big data arrays. Maybe I can test it tomorrow.

What I'm not sure about is why an _ArrayAdd function should split a string parameter and add single parts by default.

If I have a function "Add(a, b )", a user IMHO would intuitively think that "b" is added, and not transformed into other elements which then would be added. That's not intuitive.

I would suggest, that the function behaves the other way around, that is: add the parameter/object unchanged by default, and do something tricky if the user gives additional parameters to do so.